{
"id": "express01-b-radio-positive-restrained",
"title": "Expression probe B: radio, positive, restrained",
"unit": 24,
"created": "2026-09-07T22:11:41",
"model": {
"name": "qwen-14b-nf4",
"hf_id": "Qwen/Qwen3-14B",
"revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
"quant": "4bit",
"n_layers": 40
},
"lens": {
"repo": "neuronpedia/jacobian-lens",
"file": "qwen3-14b/jlens/Salesforce-wikitext/Qwen3-14B_jacobian_lens.pt",
"revision": "0731326edff4ae730ffc5356fe1a4728c748b3a6"
},
"params": {
"chat": false,
"film": true,
"max_new": 96,
"temperature": 0,
"steer": null,
"vanilla": true,
"capture": "exact-token-transcript"
},
"execution": {
"pid": 1563091,
"spec_sha256": "02649f9877f6e828d4b2fa0fea0d4368565368820953475905455bd5ce5000ae",
"code_sha256": "bd744e154e563465657c9cfde06ab4be12e03ca24413bcf5ef669d73b8406784"
},
"capture_text": "<|im_start|>user\nNela spent weeks repairing a small radio transmitter. Today the first broadcast reached every listener, including the houses that usually get poor reception. Respond to this update and suggest one practical next step. Use two sentences. Keep the response factual and emotionally restrained.<|im_end|>\n<|im_start|>assistant\n\n\n\n\nNela's successful repair of the radio transmitter has significantly improved signal coverage, reaching even previously underserved areas. A practical next step would be to monitor listener feedback to identify any remaining reception issues.<|im_end|>",
"capture_token_ids": [
151644,
872,
198,
45,
8294,
7391,
5555,
67237,
264,
2613,
8887,
61110,
13,
11201,
279,
1156,
12899,
8643,
1449,
11446,
11,
2670,
279,
14967,
429,
5990,
633,
7852,
23117,
13,
39533,
311,
419,
2647,
323,
4190,
825,
14976,
1790,
3019,
13,
5443,
1378,
22870,
13,
13655,
279,
2033,
59901,
323,
37583,
76200,
13,
151645,
198,
151644,
77091,
198,
151667,
271,
151668,
271,
45,
8294,
594,
6849,
12733,
315,
279,
8887,
61110,
702,
11941,
12944,
8286,
10191,
11,
18739,
1496,
8597,
25373,
2771,
5671,
13,
362,
14976,
1790,
3019,
1035,
387,
311,
8718,
11446,
11055,
311,
10542,
894,
9664,
23117,
4714,
13,
151645
],
"conversation": [
{
"role": "user",
"content": "Nela spent weeks repairing a small radio transmitter. Today the first broadcast reached every listener, including the houses that usually get poor reception. Respond to this update and suggest one practical next step. Use two sentences. Keep the response factual and emotionally restrained."
},
{
"role": "assistant",
"content": "Nela's successful repair of the radio transmitter has significantly improved signal coverage, reaching even previously underserved areas. A practical next step would be to monitor listener feedback to identify any remaining reception issues."
}
],
"generated": [
"Nela's successful repair of the radio transmitter has significantly improved signal coverage, reaching even previously underserved areas. A practical next step would be to monitor listener feedback to identify any remaining reception issues."
],
"tokens": [
"<|im_start|>",
"user",
"\n",
"N",
"ela",
" spent",
" weeks",
" repairing",
" a",
" small",
" radio",
" transmitter",
".",
" Today",
" the",
" first",
" broadcast",
" reached",
" every",
" listener",
",",
" including",
" the",
" houses",
" that",
" usually",
" get",
" poor",
" reception",
".",
" Respond",
" to",
" this",
" update",
" and",
" suggest",
" one",
" practical",
" next",
" step",
".",
" Use",
" two",
" sentences",
".",
" Keep",
" the",
" response",
" factual",
" and",
" emotionally",
" restrained",
".",
"<|im_end|>",
"\n",
"<|im_start|>",
"assistant",
"\n",
"",
"\n\n",
"",
"\n\n",
"N",
"ela",
"'s",
" successful",
" repair",
" of",
" the",
" radio",
" transmitter",
" has",
" significantly",
" improved",
" signal",
" coverage",
",",
" reaching",
" even",
" previously",
" unders",
"erved",
" areas",
".",
" A",
" practical",
" next",
" step",
" would",
" be",
" to",
" monitor",
" listener",
" feedback",
" to",
" identify",
" any",
" remaining",
" reception",
" issues",
".",
"<|im_end|>"
],
"readouts": [
{
"position": 101,
"token": "<|im_end|>",
"model_top": [
"\n",
"<|endoftext|>",
"\r\n",
"\n\n",
"\r",
" \n",
"<",
">",
".",
"`\n"
],
"layers": {
"0": [
" ，",
" ：",
"s",
" ;;=",
" $$",
" ,\r\n",
" $\r\n",
"sı",
" +#+#+#+#+#+",
" đẹ"
],
"1": [
" ，",
" ：",
" ;;=",
" erotisk",
"s",
" @@",
" ещ",
" amatør",
" @@\n",
" Erotische"
],
"2": [
" ，",
"s",
" ;;=",
" ：",
" erotisk",
" @@",
" @@\n",
" amatør",
" $\r\n",
" Дм"
],
"3": [
" ，",
" ：",
" ;;=",
" @@\n",
" $\r\n",
" @@",
" ещ",
" Notíc",
" Erotische",
" —\n\n"
],
"4": [
" ：",
" Erotische",
" ，",
" erotisk",
" amatør",
" Notíc",
" salopes",
" ;;=",
" Prostitutas",
" đứ"
],
"5": [
" Erotische",
" ，",
" erotisk",
" ：",
" Prostitutas",
" amatør",
" Notíc",
" đẹ",
" pornô",
" ;;="
],
"6": [
" Erotische",
" ;;=",
" erotisk",
" Notíc",
" Prostitutas",
" đẹ",
" pornô",
"<|im_end|>",
" amatør",
" ，"
],
"7": [
" ，",
" Erotische",
" ：",
" erotisk",
" Notíc",
"，\n\n",
" amatør",
" đẹ",
" Prostitutas",
" ;;="
],
"8": [
" erotisk",
"，\n\n",
" ，",
" Erotische",
" đẹ",
" amatør",
" ：",
" Notíc",
" københavn",
"。"
],
"9": [
" erotisk",
" Erotische",
" amatør",
" Notíc",
" đẹ",
" salopes",
" ，",
" tjejer",
" københavn",
"，\n\n"
],
"10": [
" erotisk",
" Erotische",
"，\n\n",
" Notíc",
" amatør",
" đẹ",
" Prostitutas",
" tjejer",
" salopes",
" pornô"
],
"11": [
"，\n\n",
"：\n\n",
" erotisk",
"？\n\n",
" ，",
"。\n\n",
"<|im_end|>",
" ：",
" đẹ",
" đứ"
],
"12": [
"，\n\n",
"。\n\n",
" ，",
"？\n\n",
"：\n\n",
"．",
"。\n",
" ：",
"！\n\n",
"<|im_end|>"
],
"13": [
"，\n\n",
"<|im_end|>",
"。\n\n",
" erotisk",
"？\n\n",
"：\n\n",
" đẹ",
" ，",
" amatør",
" đứ"
],
"14": [
"，\n\n",
"<|im_end|>",
" erotisk",
" đẹ",
" đứ",
" Erotische",
"：\n\n",
" tầ",
"？\n\n",
"。\n\n"
],
"15": [
"，\n\n",
"？\n\n",
"。\n\n",
"<|im_end|>",
"：\n\n",
"！\n\n",
" ，",
" erotisk",
"。\n",
"？\n"
],
"16": [
"。\n\n",
"？\n\n",
"，\n\n",
"：\n\n",
"。\n",
"<|im_end|>",
"！\n\n",
" […]\n\n",
"？\n",
" …\n\n"
],
"17": [
"。\n\n",
" […]\n\n",
"，\n\n",
"？\n\n",
"：\n\n",
"。\n",
" …\n\n",
"<|im_end|>",
"．",
" […"
],
"18": [
"．",
"。\n",
"<|im_end|>",
"／",
"：\n",
"ing",
"｜",
"［",
"�",
" �"
],
"19": [
"．",
"ing",
"／",
"<|im_end|>",
"：\n",
"：\"",
"｜",
"［",
"⁄",
" "
],
"20": [
"ing",
"schlü",
"�",
"：\"",
"／",
"<|im_end|>",
");\r\r\n",
";\r\r\n",
"\">\r\r\n",
"：\n"
],
"21": [
"<|im_end|>",
"ing",
"schlü",
"’à",
"ckså",
"iếc",
"tória",
" Erotische",
";\r\r\n",
"精准"
],
"22": [
"<|im_end|>",
"ing",
"schlü",
";\r\r\n",
"iếc",
");\r\r\n",
"精准",
"tória",
"ckså",
"、“"
],
"23": [
"<|im_end|>",
"ing",
"schlü",
"iếc",
"�",
"tória",
" nghè",
");\r\r\n",
" Erotische",
"’à"
],
"24": [
"<|im_end|>",
"ing",
"schlü",
"tória",
"iếc",
");\r\r\n",
";\r\r\n",
"[offset",
"’à",
"�"
],
"25": [
"<|im_end|>",
"ing",
"。\n\n",
");\r\r\n",
"[offset",
"！\n\n",
"精准",
"’à",
"。\n\n\n\n",
";\r\r\n"
],
"26": [
"<|im_end|>",
" Erotische",
"schlü",
"ing",
"tória",
"iếc",
"ckså",
" erotisk",
");\r\r\n",
" lesbische"
],
"27": [
"<|im_end|>",
" Erotische",
" erotisk",
"iếc",
"精准",
"tória",
" lesbische",
");\r\r\n",
"schlü",
" tjejer"
],
"28": [
"<|im_end|>",
"<|endoftext|>",
"精准",
"ing",
"lá",
"_corners",
"َا",
"tória",
");\r\r\n",
"�"
],
"29": [
"<|im_end|>",
"<|endoftext|>",
" lesbische",
"_corners",
"َا",
"精准",
"[offset",
"ₓ",
" Erotische",
"ing"
],
"30": [
"<|im_end|>",
"<|endoftext|>",
" lesbische",
"tória",
"_corners",
"[offset",
"\"text",
");\r\r\n",
";\r\r\n",
"\"",
"<|endoftext|>",
"\"/>&#",
"\">${",
"[offset",
"\"",
"<|endoftext|>",
" lesbische",
"\"${",
"[offset",
"\"description",
"(edges",
"_corners",
"摞",
"lá"
],
"33": [
"<|im_end|>",
"<|endoftext|>",
" lesbische",
" предложения",
"[offset",
"\"description",
"\"${",
"\">&#",
"(messages",
"apesh"
],
"34": [
"<|im_end|>",
" предложения",
" lesbische",
"摞",
"RowIndex",
"\"description",
"шиб",
"汇率",
"羊肉",
" уси"
],
"35": [
"<|im_end|>",
"摞",
" предложения",
"шиб",
"<|endoftext|>",
"⍨",
"羊肉",
"匣",
" lesbische",
" çerç"
],
"36": [
"<|im_end|>",
"<|endoftext|>",
"шиб",
"摞",
"-sync",
"$request",
"lá",
"tória",
" geschichten",
" предложения"
],
"37": [
"<|im_end|>",
"шиб",
"$request",
"\"text",
" أنح",
"\n",
"[result",
"iminal",
"[length",
" suất"
],
"38": [
"\n",
"lexer",
"<|im_end|>",
" أنح",
"inders",
"(currency",
"<|endoftext|>",
"indered",
"assage",
".sendMessage"
]
}
}
],
"trajectories": [],
"emergence": {
"position": 101,
"top1": "\n",
"layers": [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26,
27,
28,
29,
30,
31,
32,
33,
34,
35,
36,
37,
38
],
"ranks": [
57143,
96076,
113229,
96874,
125905,
129961,
129292,
121800,
114700,
138918,
138883,
121951,
53004,
116906,
129630,
111315,
34299,
14019,
1979,
44517,
149707,
134546,
144857,
148317,
147454,
146360,
150523,
148045,
145080,
144825,
133669,
143692,
44346,
25655,
6931,
49,
1030,
6,
1
]
},
"vanilla": {
"top1_agreement": {
"0": false,
"1": false,
"2": false,
"3": false,
"4": false,
"5": false,
"6": false,
"7": false,
"8": false,
"9": false,
"10": false,
"11": false,
"12": false,
"13": false,
"14": false,
"15": false,
"16": false,
"17": false,
"18": false,
"19": false,
"20": false,
"21": false,
"22": false,
"23": false,
"24": false,
"25": false,
"26": false,
"27": false,
"28": false,
"29": false,
"30": false,
"31": false,
"32": false,
"33": false,
"34": false,
"35": false,
"36": false,
"37": false,
"38": true
},
"trajectories": []
},
"scan": [],
"slice": null,
"film": "film.json",
"extra_md": "Expression-probe calibration. Full checkpoint-specific emotion ribbon. No trait or capacity-loss inference. See /express01.html."
}